HIV pre-exposure prophylaxis (PrEP) is a proven HIV prevention strategy in which HIV negative persons use antiretroviral therapy prior to potential HIV exposure. Several clinical trial have shown that truvada(r) is effective; however, there has been substantial variation in observed effectiveness across the studies - a difference that seems to be explained by varying adherence. These differences in adherence across studies have complicated the interpretation of HIV efficacy and safety (e.g. measures of bone mineral density and renal function) and the PrEP field needs objective measures to assess adherence and an approach to estimate the effect of truvada if taken daily. Pharmacologic measures of truvada dosing offer a promising, objective assessment of adherence assessment; however, there are several possible measures. To date, it is unclear which of these is the most powerful in predicting adherence. We propose a series of analyses utilizing two kinds of available datasets - (i) iPrEx OLE which has clinical outcomes including HIV infection, renal function, and bone mineral density in 1225 PrEP users with the full spectrum of drug level measured and (ii) three pharmacology datasets of directly observed truvada therapy. The pharmacology studies assigned volunteers to take truvada or tenofovir under prescribed schedules (e.g., daily, 4 pills/week, 2 pills/week) with directly observed dosing and measured various types of drug levels. We plan to compare how drug levels, of various types, vary across dosing patterns in the pharmacology studies. We will assess which types appear to be best in differentiating between dosing scheduled and develop an approach to predicting dosing patterns based on drug levels. We will then apply these classifications to the clinical dataset (iPrEx OLE). With this classification, we will estimate the risk of HIV infection, the changes in bone density and kidney function across potential dosing patterns (e.g., if truvada were taking daily, or as 2 or 4 pills/week). This work will establish the drug test or tests that are the strongest correlates of truvada use and estimate the clinical impacts of truvada under a variety of potential dosing patterns.